Study design is an incomplete randomized block design. Fixed effects are maturity (covariate/continuous variable) and type (A, B, or C). There are 20 total samples with 7 of two types and 6 of one type. Samples were put into blocks with at least one of each type in the blocks. Two Blocks were analyzed over 4 runs, so Block is nested within Run. The model is Y = Maturity + Type + Block-within-Run, with Block-within-Run as a random effect. I accidentally coded it as 1|(Block:Run) instead of (1|Block:Run) and got an error:

Error: number of observations (=40) <= number of random effects (=48) for term (kernel_dm * type + 1 | (block:run)); the random-effects parameters and the residual variance (or scale parameter) are probably unidentifiable

So I just wondered what the difference was and what the | in this case is actually doing?